The purpose of this analysis document is to ensure the reproducability of the results by guiding the reader through the random forest analysis of the factors associated with the health of western redcedar.
Root data were shared by citizen scientists in the Wester Redcedar Dieback Map project on iNaturalist.
All of the data used in the below analyses are described in the Data Wrangle folder.
The data used in the below visualizations are described in the Data Wrangle folder.
All tree health categories
## # A tibble: 11 x 2
## # Groups: field.tree.canopy.symptoms [11]
## field.tree.canopy.symptoms n
## <fct> <int>
## 1 Branch Dieback or 'Flagging' 19
## 2 Browning Canopy 19
## 3 Extra Cone Crop 2
## 4 Healthy 403
## 5 Multiple Symptoms (please list in Notes) 17
## 6 New Dead Top (red or brown needles still attached) 33
## 7 Old Dead Top (needles already gone) 83
## 8 Other (please describe in Notes) 8
## 9 Thinning Canopy 118
## 10 Tree is dead 37
## 11 Yellowing Canopy 10
We need to filter the data to only include response and explanatory variables we’re interested in. For example, whether a sound clip was included in the iNat data is not important.
We also need to remove other response variables like “field.percent.canopy.affected….” so it is not used as a predictor for tree health.
Note it might be interesting to know if the user was an important factor in predicting if the tree is healthy/unhealthy.
There are also a number of factors that should probably be removed because they may be biasing the data. For example, only trees with the ‘other factor’ question may only be answered for unhealthy trees. We need to think about this a bit more.
We continue to get the below error, but were able to work around it by imputing the data.
Error in randomForest.default(m, y, …) : Need at least two classes to do classification.
To impute the data we have to remove factors with >53 levels.
The below code lists the number of levels for the variables that are factors.
Imputed data table
## ntree OOB 1 2 3 4 5 6 7 8 9 10 11
## 300: 46.46% 94.74% 94.74%100.00% 16.38% 88.24% 87.88% 74.70%100.00% 72.03% 94.59%100.00%
## ntree OOB 1 2 3 4 5 6 7 8 9 10 11
## 300: 45.66% 94.74% 94.74%100.00% 15.14% 82.35% 87.88% 74.70%100.00% 73.73% 89.19%100.00%
## ntree OOB 1 2 3 4 5 6 7 8 9 10 11
## 300: 46.33% 94.74% 94.74%100.00% 15.14% 76.47% 87.88% 75.90%100.00% 75.42% 97.30%100.00%
## ntree OOB 1 2 3 4 5 6 7 8 9 10 11
## 300: 45.66% 94.74%100.00%100.00% 14.89% 82.35% 87.88% 73.49%100.00% 72.88% 94.59%100.00%
## ntree OOB 1 2 3 4 5 6 7 8 9 10 11
## 300: 45.39% 94.74% 94.74%100.00% 15.14% 82.35% 87.88% 72.29%100.00% 73.73% 89.19%100.00%
## ntree OOB 1 2 3 4 5 6 7 8 9 10 11
## 300: 46.06%100.00% 94.74%100.00% 15.63% 70.59% 87.88% 75.90%100.00% 74.58% 89.19%100.00%
##
## Call:
## randomForest(formula = field.tree.canopy.symptoms ~ ., data = training, ntree = 2001, importance = TRUE, proximity = TRUE, na.action = na.omit)
## Type of random forest: classification
## Number of trees: 2001
## No. of variables tried at each split: 23
##
## OOB estimate of error rate: 43.85%
## Confusion matrix:
## Branch Dieback or 'Flagging'
## Branch Dieback or 'Flagging' 0
## Browning Canopy 0
## Extra Cone Crop 0
## Healthy 2
## Multiple Symptoms (please list in Notes) 0
## New Dead Top (red or brown needles still attached) 0
## Old Dead Top (needles already gone) 1
## Other (please describe in Notes) 0
## Thinning Canopy 1
## Tree is dead 0
## Yellowing Canopy 0
## Browning Canopy
## Branch Dieback or 'Flagging' 0
## Browning Canopy 0
## Extra Cone Crop 0
## Healthy 1
## Multiple Symptoms (please list in Notes) 1
## New Dead Top (red or brown needles still attached) 2
## Old Dead Top (needles already gone) 0
## Other (please describe in Notes) 0
## Thinning Canopy 0
## Tree is dead 2
## Yellowing Canopy 0
## Extra Cone Crop Healthy
## Branch Dieback or 'Flagging' 0 9
## Browning Canopy 0 5
## Extra Cone Crop 0 1
## Healthy 0 267
## Multiple Symptoms (please list in Notes) 0 6
## New Dead Top (red or brown needles still attached) 0 12
## Old Dead Top (needles already gone) 0 25
## Other (please describe in Notes) 0 7
## Thinning Canopy 0 39
## Tree is dead 0 12
## Yellowing Canopy 0 5
## Multiple Symptoms (please list in Notes)
## Branch Dieback or 'Flagging' 0
## Browning Canopy 0
## Extra Cone Crop 0
## Healthy 1
## Multiple Symptoms (please list in Notes) 4
## New Dead Top (red or brown needles still attached) 0
## Old Dead Top (needles already gone) 1
## Other (please describe in Notes) 0
## Thinning Canopy 1
## Tree is dead 0
## Yellowing Canopy 0
## New Dead Top (red or brown needles still attached)
## Branch Dieback or 'Flagging' 0
## Browning Canopy 2
## Extra Cone Crop 0
## Healthy 5
## Multiple Symptoms (please list in Notes) 0
## New Dead Top (red or brown needles still attached) 4
## Old Dead Top (needles already gone) 1
## Other (please describe in Notes) 0
## Thinning Canopy 3
## Tree is dead 4
## Yellowing Canopy 1
## Old Dead Top (needles already gone)
## Branch Dieback or 'Flagging' 3
## Browning Canopy 2
## Extra Cone Crop 1
## Healthy 10
## Multiple Symptoms (please list in Notes) 1
## New Dead Top (red or brown needles still attached) 1
## Old Dead Top (needles already gone) 20
## Other (please describe in Notes) 0
## Thinning Canopy 18
## Tree is dead 6
## Yellowing Canopy 0
## Other (please describe in Notes)
## Branch Dieback or 'Flagging' 0
## Browning Canopy 0
## Extra Cone Crop 0
## Healthy 0
## Multiple Symptoms (please list in Notes) 0
## New Dead Top (red or brown needles still attached) 0
## Old Dead Top (needles already gone) 0
## Other (please describe in Notes) 0
## Thinning Canopy 0
## Tree is dead 0
## Yellowing Canopy 0
## Thinning Canopy Tree is dead
## Branch Dieback or 'Flagging' 2 0
## Browning Canopy 0 2
## Extra Cone Crop 0 0
## Healthy 12 6
## Multiple Symptoms (please list in Notes) 1 0
## New Dead Top (red or brown needles still attached) 3 2
## Old Dead Top (needles already gone) 16 1
## Other (please describe in Notes) 0 0
## Thinning Canopy 19 1
## Tree is dead 3 1
## Yellowing Canopy 2 0
## Yellowing Canopy class.error
## Branch Dieback or 'Flagging' 0 1.0000000
## Browning Canopy 0 1.0000000
## Extra Cone Crop 0 1.0000000
## Healthy 1 0.1245902
## Multiple Symptoms (please list in Notes) 0 0.6923077
## New Dead Top (red or brown needles still attached) 1 0.8400000
## Old Dead Top (needles already gone) 0 0.6923077
## Other (please describe in Notes) 0 1.0000000
## Thinning Canopy 1 0.7710843
## Tree is dead 0 0.9642857
## Yellowing Canopy 0 1.0000000
Selected tree health categories
## # A tibble: 5 x 2
## # Groups: field.tree.canopy.symptoms [5]
## field.tree.canopy.symptoms n
## <fct> <int>
## 1 Healthy 403
## 2 New Dead Top (red or brown needles still attached) 33
## 3 Old Dead Top (needles already gone) 83
## 4 Thinning Canopy 118
## 5 Tree is dead 37
##
## Call:
## randomForest(formula = field.tree.canopy.symptoms ~ ., data = training, ntree = 2001, importance = TRUE, proximity = TRUE, na.action = na.omit)
## Type of random forest: classification
## Number of trees: 2001
## No. of variables tried at each split: 23
##
## OOB estimate of error rate: 40%
## Confusion matrix:
## Healthy
## Healthy 269
## New Dead Top (red or brown needles still attached) 15
## Old Dead Top (needles already gone) 29
## Thinning Canopy 51
## Tree is dead 10
## New Dead Top (red or brown needles still attached)
## Healthy 4
## New Dead Top (red or brown needles still attached) 2
## Old Dead Top (needles already gone) 3
## Thinning Canopy 2
## Tree is dead 2
## Old Dead Top (needles already gone)
## Healthy 12
## New Dead Top (red or brown needles still attached) 4
## Old Dead Top (needles already gone) 15
## Thinning Canopy 13
## Tree is dead 2
## Thinning Canopy Tree is dead
## Healthy 19 3
## New Dead Top (red or brown needles still attached) 3 1
## Old Dead Top (needles already gone) 15 2
## Thinning Canopy 15 5
## Tree is dead 7 2
## class.error
## Healthy 0.1237785
## New Dead Top (red or brown needles still attached) 0.9200000
## Old Dead Top (needles already gone) 0.7656250
## Thinning Canopy 0.8255814
## Tree is dead 0.9130435
Binary tree health categories
## # A tibble: 2 x 2
## # Groups: field.tree.canopy.symptoms [2]
## field.tree.canopy.symptoms n
## <fct> <int>
## 1 Healthy 403
## 2 Unhealthy 346
##
## Call:
## randomForest(formula = field.tree.canopy.symptoms ~ ., data = training, ntree = 2001, importance = TRUE, proximity = TRUE, na.action = na.omit)
## Type of random forest: classification
## Number of trees: 2001
## No. of variables tried at each split: 23
##
## OOB estimate of error rate: 26.92%
## Confusion matrix:
## Healthy Unhealthy class.error
## Healthy 222 81 0.2673267
## Unhealthy 70 188 0.2713178
## left daughter right daughter split var
## 1 2 3 NFFD_at
## 2 4 5 norm_Eref08
## 3 6 7 PPT05
## 4 8 9 component_taxgrtgroup
## 5 10 11 DD_18_01
## 6 0 0 <NA>
## 7 12 13 RH_at
## 8 14 15 field.optional...slope.position
## 9 0 0 <NA>
## 10 16 17 norm_CMI03
## 11 18 19 norm_NFFD_at
## 12 20 21 norm_CMI12
## 13 22 23 norm_DD5_02
## 14 24 25 field.optional...site.location.description
## 15 26 27 component_totalsub_r
## 16 28 29 norm_DD_0
## 17 30 31 DD5_05
## 18 0 0 <NA>
## 19 32 33 component_weg
## 20 34 35 muaggatt_slopegradwta
## 21 36 37 field.optional...site.hydrology
## 22 38 39 PPT_at
## 23 40 41 PNWSRTMDEM.x
## 24 42 43 CMI06
## 25 0 0 <NA>
## 26 44 45 component_totalsub_h
## 27 46 47 norm_PPT09
## 28 0 0 <NA>
## 29 48 49 Tmin10
## 30 50 51 norm_Tmin04
## 31 0 0 <NA>
## 32 52 53 component_ffd_l
## 33 0 0 <NA>
## 34 54 55 CMI04
## 35 56 57 field.optional...tree.size
## 36 58 59 Tave12
## 37 0 0 <NA>
## 38 0 0 <NA>
## 39 60 61 CMI_at
## 40 0 0 <NA>
## 41 62 63 norm_Tmax11
## 42 0 0 <NA>
## 43 0 0 <NA>
## 44 64 65 CMI09
## 45 66 67 PPT06
## 46 0 0 <NA>
## 47 68 69 PPT01
## 48 0 0 <NA>
## 49 70 71 norm_CMD
## 50 0 0 <NA>
## 51 72 73 norm_Tmax_wt
## 52 0 0 <NA>
## 53 0 0 <NA>
## 54 74 75 Profilecurv.x
## 55 76 77 eFFP
## 56 78 79 component_slopelenusle_r
## 57 0 0 <NA>
## 58 0 0 <NA>
## 59 0 0 <NA>
## 60 0 0 <NA>
## 61 0 0 <NA>
## 62 0 0 <NA>
## 63 0 0 <NA>
## 64 80 81 DD5_04
## 65 82 83 RH_sp
## 66 84 85 norm_Eref05
## 67 86 87 norm_Tmin08
## 68 0 0 <NA>
## 69 88 89 Tangentialc.x
## 70 0 0 <NA>
## 71 90 91 DD_18_05
## 72 0 0 <NA>
## 73 92 93 norm_CMI_at
## 74 94 95 Eref_sm
## 75 96 97 component_totalsub_r
## 76 0 0 <NA>
## 77 0 0 <NA>
## 78 0 0 <NA>
## 79 0 0 <NA>
## 80 98 99 Tmax_at
## 81 100 101 Tmax08
## 82 102 103 DD_18_02
## 83 104 105 norm_DD5_11
## 84 0 0 <NA>
## 85 0 0 <NA>
## 86 106 107 field.optional...tree.size
## 87 0 0 <NA>
## 88 108 109 norm_NFFD_wt
## 89 0 0 <NA>
## 90 110 111 CMI11
## 91 0 0 <NA>
## 92 112 113 component_corsteel
## 93 0 0 <NA>
## 94 114 115 norm_CMI_wt
## 95 116 117 component_irrcapcl
## 96 0 0 <NA>
## 97 0 0 <NA>
## 98 118 119 DD5_09
## 99 120 121 PPT_wt
## 100 0 0 <NA>
## 101 0 0 <NA>
## 102 0 0 <NA>
## 103 0 0 <NA>
## 104 0 0 <NA>
## 105 122 123 norm_Tmin09
## 106 0 0 <NA>
## 107 124 125 norm_DD5_07
## 108 0 0 <NA>
## 109 126 127 norm_Eref_at
## 110 0 0 <NA>
## 111 0 0 <NA>
## 112 128 129 Eref04
## 113 0 0 <NA>
## 114 130 131 Tmax10
## 115 0 0 <NA>
## 116 0 0 <NA>
## 117 132 133 norm_SHM
## 118 0 0 <NA>
## 119 134 135 PPT02
## 120 0 0 <NA>
## 121 0 0 <NA>
## 122 136 137 norm_CMI_sm
## 123 0 0 <NA>
## 124 0 0 <NA>
## 125 0 0 <NA>
## 126 0 0 <NA>
## 127 0 0 <NA>
## 128 138 139 Tave06
## 129 0 0 <NA>
## 130 140 141 norm_DD5_sp
## 131 0 0 <NA>
## 132 142 143 muaggatt_brockdepmin
## 133 0 0 <NA>
## 134 0 0 <NA>
## 135 0 0 <NA>
## 136 0 0 <NA>
## 137 0 0 <NA>
## 138 144 145 DD5_09
## 139 146 147 norm_CMD09
## 140 0 0 <NA>
## 141 148 149 component_albedodry_l
## 142 0 0 <NA>
## 143 0 0 <NA>
## 144 0 0 <NA>
## 145 0 0 <NA>
## 146 150 151 CMI08
## 147 152 153 component_totalsub_r
## 148 154 155 CMI01
## 149 0 0 <NA>
## 150 156 157 norm_DD18
## 151 0 0 <NA>
## 152 0 0 <NA>
## 153 0 0 <NA>
## 154 0 0 <NA>
## 155 0 0 <NA>
## 156 0 0 <NA>
## 157 0 0 <NA>
## split point status prediction
## 1 8.450000e+01 1 <NA>
## 2 1.125000e+02 1 <NA>
## 3 4.800000e+01 1 <NA>
## 4 2.001079e+14 1 <NA>
## 5 4.460000e+02 1 <NA>
## 6 0.000000e+00 -1 Unhealthy
## 7 7.350000e+01 1 <NA>
## 8 1.304000e+03 1 <NA>
## 9 0.000000e+00 -1 Healthy
## 10 8.085000e+00 1 <NA>
## 11 6.150000e+01 1 <NA>
## 12 1.655000e+01 1 <NA>
## 13 5.650000e+01 1 <NA>
## 14 2.400000e+01 1 <NA>
## 15 7.499997e+01 1 <NA>
## 16 6.750000e+01 1 <NA>
## 17 2.945000e+02 1 <NA>
## 18 0.000000e+00 -1 Unhealthy
## 19 6.500000e+00 1 <NA>
## 20 3.400000e+01 1 <NA>
## 21 4.000000e+00 1 <NA>
## 22 3.350000e+02 1 <NA>
## 23 2.700000e+01 1 <NA>
## 24 -2.935000e+00 1 <NA>
## 25 0.000000e+00 -1 Unhealthy
## 26 3.599999e+01 1 <NA>
## 27 4.450000e+01 1 <NA>
## 28 0.000000e+00 -1 Unhealthy
## 29 6.950000e+00 1 <NA>
## 30 3.450000e+00 1 <NA>
## 31 0.000000e+00 -1 Healthy
## 32 9.500000e+01 1 <NA>
## 33 0.000000e+00 -1 Unhealthy
## 34 3.955000e+00 1 <NA>
## 35 6.000000e+00 1 <NA>
## 36 4.550000e+00 1 <NA>
## 37 0.000000e+00 -1 Healthy
## 38 0.000000e+00 -1 Healthy
## 39 1.865000e+01 1 <NA>
## 40 0.000000e+00 -1 Healthy
## 41 1.095000e+01 1 <NA>
## 42 0.000000e+00 -1 Unhealthy
## 43 0.000000e+00 -1 Healthy
## 44 5.335000e+00 1 <NA>
## 45 5.250000e+01 1 <NA>
## 46 0.000000e+00 -1 Unhealthy
## 47 1.310000e+02 1 <NA>
## 48 0.000000e+00 -1 Unhealthy
## 49 3.255000e+02 1 <NA>
## 50 0.000000e+00 -1 Healthy
## 51 7.250000e+00 1 <NA>
## 52 0.000000e+00 -1 Unhealthy
## 53 0.000000e+00 -1 Healthy
## 54 4.400442e-03 1 <NA>
## 55 3.295000e+02 1 <NA>
## 56 8.620339e+01 1 <NA>
## 57 0.000000e+00 -1 Healthy
## 58 0.000000e+00 -1 Unhealthy
## 59 0.000000e+00 -1 Healthy
## 60 0.000000e+00 -1 Unhealthy
## 61 0.000000e+00 -1 Healthy
## 62 0.000000e+00 -1 Unhealthy
## 63 0.000000e+00 -1 Healthy
## 64 1.395000e+02 1 <NA>
## 65 6.850000e+01 1 <NA>
## 66 1.000000e+02 1 <NA>
## 67 1.215000e+01 1 <NA>
## 68 0.000000e+00 -1 Unhealthy
## 69 -5.707497e-04 1 <NA>
## 70 0.000000e+00 -1 Unhealthy
## 71 1.340000e+02 1 <NA>
## 72 0.000000e+00 -1 Healthy
## 73 5.258000e+01 1 <NA>
## 74 4.095000e+02 1 <NA>
## 75 7.499997e+01 1 <NA>
## 76 0.000000e+00 -1 Unhealthy
## 77 0.000000e+00 -1 Healthy
## 78 0.000000e+00 -1 Healthy
## 79 0.000000e+00 -1 Unhealthy
## 80 1.490000e+01 1 <NA>
## 81 2.490000e+01 1 <NA>
## 82 4.135000e+02 1 <NA>
## 83 2.250000e+01 1 <NA>
## 84 0.000000e+00 -1 Unhealthy
## 85 0.000000e+00 -1 Healthy
## 86 1.400000e+01 1 <NA>
## 87 0.000000e+00 -1 Healthy
## 88 5.750000e+01 1 <NA>
## 89 0.000000e+00 -1 Healthy
## 90 1.215500e+01 1 <NA>
## 91 0.000000e+00 -1 Healthy
## 92 2.000000e+00 1 <NA>
## 93 0.000000e+00 -1 Healthy
## 94 4.163500e+01 1 <NA>
## 95 2.927692e+00 1 <NA>
## 96 0.000000e+00 -1 Unhealthy
## 97 0.000000e+00 -1 Healthy
## 98 3.065000e+02 1 <NA>
## 99 4.200000e+02 1 <NA>
## 100 0.000000e+00 -1 Unhealthy
## 101 0.000000e+00 -1 Healthy
## 102 0.000000e+00 -1 Healthy
## 103 0.000000e+00 -1 Unhealthy
## 104 0.000000e+00 -1 Healthy
## 105 8.450000e+00 1 <NA>
## 106 0.000000e+00 -1 Unhealthy
## 107 3.885000e+02 1 <NA>
## 108 0.000000e+00 -1 Healthy
## 109 1.275000e+02 1 <NA>
## 110 0.000000e+00 -1 Healthy
## 111 0.000000e+00 -1 Unhealthy
## 112 7.750000e+01 1 <NA>
## 113 0.000000e+00 -1 Unhealthy
## 114 1.570000e+01 1 <NA>
## 115 0.000000e+00 -1 Unhealthy
## 116 0.000000e+00 -1 Unhealthy
## 117 1.127500e+02 1 <NA>
## 118 0.000000e+00 -1 Healthy
## 119 1.820000e+02 1 <NA>
## 120 0.000000e+00 -1 Healthy
## 121 0.000000e+00 -1 Unhealthy
## 122 1.345000e+00 1 <NA>
## 123 0.000000e+00 -1 Unhealthy
## 124 0.000000e+00 -1 Healthy
## 125 0.000000e+00 -1 Unhealthy
## 126 0.000000e+00 -1 Unhealthy
## 127 0.000000e+00 -1 Healthy
## 128 1.555000e+01 1 <NA>
## 129 0.000000e+00 -1 Healthy
## 130 5.305000e+02 1 <NA>
## 131 0.000000e+00 -1 Healthy
## 132 1.363855e+02 1 <NA>
## 133 0.000000e+00 -1 Healthy
## 134 0.000000e+00 -1 Unhealthy
## 135 0.000000e+00 -1 Healthy
## 136 0.000000e+00 -1 Unhealthy
## 137 0.000000e+00 -1 Healthy
## 138 3.195000e+02 1 <NA>
## 139 4.600000e+01 1 <NA>
## 140 0.000000e+00 -1 Healthy
## 141 1.889381e-01 1 <NA>
## 142 0.000000e+00 -1 Unhealthy
## 143 0.000000e+00 -1 Healthy
## 144 0.000000e+00 -1 Unhealthy
## 145 0.000000e+00 -1 Healthy
## 146 -1.161500e+01 1 <NA>
## 147 8.999996e+00 1 <NA>
## 148 1.193500e+01 1 <NA>
## 149 0.000000e+00 -1 Healthy
## 150 2.215000e+02 1 <NA>
## 151 0.000000e+00 -1 Unhealthy
## 152 0.000000e+00 -1 Unhealthy
## 153 0.000000e+00 -1 Healthy
## 154 0.000000e+00 -1 Healthy
## 155 0.000000e+00 -1 Unhealthy
## 156 0.000000e+00 -1 Healthy
## 157 0.000000e+00 -1 Unhealthy
Fit a single recursive partitioning or classification tree. Followed instructions from this youtube video.
Below is an example of one of the trees included in the random forest.
Error in randomForest.default(m, y, …) : Need at least two classes to do classification.
I may be misunderstanding this error, but I think it is referring to the response variable?
The documentation here describes the error prompt when: if (classRF && !addclass && length(unique(y)) < 2) stop(“Need at least two classes to do classification.”)
It is possible some of the NA or -9999 values are causing issues.
We can try imputing the data, however this requires us to remove columns with more than 53 factors, which probably makes sense anyway.
Removing factors with more than 53 levels didn’t resolve the error from the randomForest command, but it did allow us to use the rfImpute command to impute our data.
Wow it actually worked if the data is imputed.